Fig. 1
From: Inferring Crohn’s disease association from exome sequences by integrating biological knowledge
Overview of NMTF model and constraints for distinguishing disease exome and identifying disease-associated genes. a Exome sequences are represented as the matrix V encoding the occurrences of amino acid substitutions. V is approximated by the multiplication of three different matrices P, W, and H, which represent pathogenicity of mutations, disease-associated genes, and disease status of exomes, respectively. b To obtain a biologically meaningful solution, three kinds of biological knowledge, such as pathogenicity prediction, known annotation of disease genes, and functional interaction between genes, are integrated as constraints